Digital Photogrammetric Imaging Past PresentDigital Photogrammetric Imaging – Past, Present and Future

Prof. em. Armin GruenProf. em. Armin GruenInstitute of Conservation and Building Research, ETH Zurich agruen@geod.baug.ethz.ch, www.photogrammetry.ethz.ch

1

40th Ph t W k St tt t 198540th Phot. Week, Stuttgart 1985Ackermann, F.: Technology transfer – a glance back at 39 Photogrammetric WeeksPhotogrammetric Weeks„... It can be stated ... that the primary concern of mediating between science and practice embodies the general aim .... p gOne can say and hope that future meetings in the field of photogrammetry and its related disciplines continue to devote th l t thi t k ith t d “themselves to this task with engagement and success.“

Participants: 1909 -1913: 32Participants: 1909 1913: 321973 -1985: 240

Congratulations !

2

3

Courtesy Gottfried Konecny

D l t f di it l h t tDevelopment of digital photogrammetry

Early 80s: Paradigm change “From Analytical to Digital”y g g y g

Requires “fore-runner technology”:

+ Sensors

+ Computing platforms

+ Software (lacking) Software (lacking)

ContentsContents-The early years (1970-1984)

- First systems and experiences (1984-1988)First systems and experiences (1984 1988)

- Acceptance and refinement (1988-1994)

- Time of consolidation (1994-2000)

4

- Time of consolidation (1994-2000)

- Fully digital era (2000-now)

The Future ?“I never think of the future. It comes soon enough.”

Alb t Ei t iAlbert Einstein

Safe prediction:S S ll P i F t

5

Sensors: Smaller, Processing: Faster, Results: Better(?)

The Foundations

1964 - 1980: Image processing algorithms1964 1980: Image processing algorithms

1972 - 1980: Network aspects (syst. errors, blunders; reliability)

1980 1984: On line triangulation by sequential estimation (Kalman TFU Givens)1980 - 1984: On-line triangulation by sequential estimation (Kalman,TFU, Givens)

1980 - 1984: Charge Coupled Devices (CCD) and applications

B tt d Li A CCD ith 10 000 l t iButted Linear Array CCDs with 10 000 elements in a row

Framegrabbers, Workstations (Tektronix, Sun)

Image processing systems with data acquisition devices

1984: L.S. image and template matching (algorithmic basis)

6

The early years (1970 – 1984)

1972 Landsat 1, NASA, Satellite remote sensing1976 13th Congress of ISP, Helsinki

7 Analytical Plotters, Analogue → Analytical7 Analytical Plotters, Analogue Analytical1979 Mapsat (Colvocoresses, epipolar principle, along track)1980 14th Congress of ISP, Hamburg, ISP → ISPRS1981 St t (W l h&M k l t k i t i l t)1981 Stereosat (Welch&Marko, along track image triplet)1982 DPS (Hofmann, image triplet)

1984 15th Congress of ISPRS Rio de Janeiro1984 15th Congress of ISPRS, Rio de Janeiro> Introduction of digital close-range photogrammetry (CCD cameras)

1986 Launch of SPOT-1 (22 Feb)

7

Early developments in close-range photogrammetry1974 Woltring: Human motion studies with PSD Vicon system

1976 ISP Congress Helsinki: RTP awareness

1978 Pinkney/Kratky: On-line 30Hz vidicon, resection(Space Shuttle manipulator arm)

1981 Reece: SELSPOT based research (DLT, commercial)

1983 Real: Matrix camera with digital image processing in photogrammetric applications

1983 El-Hakim: Photogrammetric robot vision

1984 Haggren, ISPRS Congress Rio de Janeiro:

New vistas for industrial photogrammetry

8

9

Pinkney, Kratky, NRC, 1978: Docking maneuver for Space Shuttle Manipulator arm

F ll di it l tFully digital systems

H. Haggren, Rio de Janeiro 1984: “The present low pixel resolution h t b k d i ”has to be cracked in some way.”

HUT prototyp system:2 id i l i d l2 video cameras, special processing modules

128x128 pi, 4 bits

Accuracy: 1 : 500 – 1 : 1000Accuracy: 1 : 500 1 : 1000

Mapvision

10

“Digital Plotters”

- Sarjakoski,1981, Case,1982, Albertz, Koenig,1982, Dowman,1982,

Jaksic,1984, Kunji,1984, Gruen,1986

- 1985 “DIPS” Digital Station at IGP, ETH Zurich

KONTRON IPS 68K

2 CCD frame transfer cameras AQUA TV (384x576, 604x576 pixels)

-1986 ISPRS Commission II Symposium, Baltimore, MarylandPhot. and Remote Sensing Systems for Data Processing and Analysis

11

H d hit t f Di it l St tiHardware architecture of a Digital Station

12

DMA specifications for DSCCDMA specifications for DSCC

- Processing of stereopair; automatically, operateur-controlled

- Photometric processingPhotometric processing

- Per photo: 20'000 x 20'000 pi

- Radiometric resolution: 8 bits

- Roam Rate: 2.5 mm/Sek (200 pi/sec)

- Fast model traversing:Establishment of stereomodel 2 sec

- DHM data generation: 200 pts/sec

DHM t hi 1 i/i- DHM matching accuracy: 1 pi/image

- Superposition („Video Map“)planimetric elements, elevation data; flicker modep , ;

- Generation of local image patches (feature points) (objects, control points, tie points)

13

i d i ( )First systems and experiences (1984 -1988)1985 Burner et al. : Calibration of vidicon tubes

R l F ji t CCD b d t tReal, Fujimoto: CCD-based stereo systemHobrough, Hobrough : STEREOPSIS system (“videogrammetry“)

2 - 6 12 1985 Cannes: SPIE Conf “Computer Vision for Robots“2. 6.12. 1985, Cannes: SPIE Conf. Computer Vision for Robots

1986 El -Hakim : CCD-Stereo system based on IRI-D 256Wong, Ho : 2 GE TN250 CID cameras (toy moose)g ( y )Murai et al. : 2 linear array CCDs (2048 pi), structured lightHaggren : MAPVISION, 4 CCD cameras, synchronization,

targetted object points, 1 sec per pointtargetted object points, 1 sec per point

1986 Gruen, Beyer : 4 CCD cameras, Kontron IPS 68K,non-simultaneous, 2D testfield

1987 Gruen, Beyer, Dähler: DIPS II, Sun, Datacube boards,3D testfield, systematic errors

14

Problems with video signalProblems with video signal

15

16

Table Tennis Robot 1988Table Tennis Robot 1988 DAGM Zurich Winner, World Champion Singapore

17

D l t fDevelopment of sensorsTerrestrial Photogrammetry

1986 Canon CI 10 + JVC Monitor 0.4 Mpi SFr 5 500

1986 Canon RC 701 SLR + player/rec. 0.2 Mpi SFr 11 200

PhoWo 14-19 Sept. 1987: Towards Real-Time Photogrammetry

1989 Kontron ProgRes 3000 7.5 Mpi, 8 sec/image, SFr 50 000 + 22 000

1993 Kodak DCS 200 1.5 Mpi SFr 16 000

Today: Hasselblad & Phase One back 39 Mpi 2 sec/imageToday: Hasselblad & Phase One back 39 Mpi, 2 sec/image

ca. SFr 40 000

18

Development of digital photogrammetryDevelopment of digital photogrammetry

ISPRS Commission II Symposium Rovaniemi, July 1986:“From Analytical to Digital”From Analytical to Digital

ISPRS Intercommission Conference on “Fast processing of photogrammetric data” Interlaken 2 4 June 1987:“Fast processing of photogrammetric data”, Interlaken, 2-4 June 1987:First conference fully devoted to digital techniques- HW for Machine Vision systems (Kontron Bildanalyse GmbH)

T t A f l ti f t t hi- Transputer Arrays for real-time feature matching- Image acquisition, image quality, calibration of video cameras- Digital processing of SPOT imagery- On-line triangulation- Image matching: Multi-image, Facets stereo (FAST) vision - Automatic DTM generation on Kern DRS-11g- Förstner Operator- Mapvision system, DCCS (Helava)- Digital ortho-images array algebra

19

Digital ortho images, array algebraPanel Discussion: “Digital Photogrammetry – Quo Vadis?”PhoWo 2009: ”Future of Photogrammetry”

K. Atkinson, ISPRS Kyoto 1988:

“Th K t C C i i V th l i t f ISP“The Kyoto Congress saw Commission V, once the ugly sister of ISP,

emerge as one of the stars of the show.”

20

Acceptance and refinement (1988-1994)

• Calibration operational

• Digital aerial cameras (medium format)• Digital aerial cameras (medium format)

• High accuracy measurement algorithms/systems

F ll di it l t t d t• Fully digital automated systems

• Digital ortho-images

• CAD interface/integration, visualization

• Diversity of applications

• Industrial photogrammetry established

• Few robotic and medical applications

21

PTVPTV measurements

Software: Free web download, 45 lic. per year27/28 August 2009: COST Conference ETH Zurich

22

27/28 August 2009: COST Conference, ETH Zurich

D l t f di it l tDevelopment of digital systems

Without scientific basis

ISPRS Conference on “Digital Photogrammetric Systems”, 4 6 September 1991 TU Munich:4-6 September 1991, TU Munich:Session “Design Research” – user interface, data transfer, image handling, measurement techniquesimage handling, measurement techniques (design no scientific basis, on-the-job experimentation)What automatic – what semi-automatic?

23

3 stages of an innovative product1. Technical solution2. Acceptance by customerccep a ce by cus o e3. Strategic instrument in the market

Technology – too early/too late?„The citizen‘s problem, at bottom, is how to assess pthe things that so often come forth in the beguiling guise of blessings.“

Wi M i li W b W d V l Zü i h 1989Wirz: Marginalien zur Werbung. Werd Verlag, Zürich,1989

24

Digital WorkstationsOn the ISPRS agenda: Since 1986On the ISPRS agenda: Since 1986 (Symposium Comm. II, Baltimore)

Kern & Co., ISPRS Congress Kyoto 1988:

W d l di i ll “

25

„We yodel digitally“

ISPRS Symposium Zürich, Sept. 1990:“Close-Range Photogrammetry Meets

Machine Vision”Machine Vision”

WG V/6: Biostereometrics and Medical Imagingg g

• Applications: Face, eye, dentistry, heart, lung, knee, hip-joint, respiration, body

surface motion (human animal) sportssurface, motion (human, animal), sports

• Sensors: CT, SEM, MRI, X-ray stereo, REM stereo,

laser scanning microscopy,

tomography, radiography, ultrasound, CCDs

1989 C V

26

1989: Comm.V

WG chairmen in Lagna

i f C lid i ( 99 2000)Time of Consolidation (1994-2000)

- Large format digital aerial cameras LiDARLarge format digital aerial cameras, LiDAR- Feature and object extraction (city models, road extraction)- 1998: 15 commercial digital stations with 3D capabilities

Image matching for DSM generation- Image matching for DSM generation

- Highres satellite sensors: IKONOS, etc.

- Terrestrial photogrammetry: + High accuracy (1:10 000 to 1:100 000), industrial/eng. applications+ Fast processing, image sequences (robotics, motion analysis, PTV)+ Surface measurements, large structures ;

laser-scanners, structured light systems, g y1998: 27 commercial systems for surface reconstruction

16 body scanners and 7 face trackers+ Mobile Mapping multi-sensor concepts

27

Mobile Mapping, multi sensor concepts1999: 10 road-based MM systems

ISPRS Congress Vienna 1996 Comm V topicsISPRS Congress Vienna 1996 – Comm.V topics• Architectural Photogrammetry

photographic/hybrid 21di it l (CCD ) 11digital (CCDs) 11

• Industrial applicationsphotographic/hybrid 3di it l (CCD ) 20digital (CCDs) 20

• Moving objects 7• Moving sensors 1• Calibration 3• Depth-from-focus 1• Network design 1g• On-line triangulation 1• Size of targets 1• Matching 2Matching 2• Meas. algorithms/accuracy 1• Orientation without CPs 1• Spatial resection closed form 1

28

• Spatial resection, closed form 1• Mikroscopy 3

Al G Vi id t USA J 1998Al Gore, Vicepresident USA, January 1998:The Digital Earth: Understanding our planet in the 21 centuryI believe we need a ‘Digital Earth’: a multi-resolution three-„I believe we need a Digital Earth : a multi resolution, three

dimensional representation of the planet, into which we can embed vast quantities of geo-referenced data.”

Digital Earth Society (Beijing) Google Earth Google Earth Microsoft Virtual World, etc.

29

F ll di it l (2000 2009)Fully digital era (2000 – 2009)

19th ISPRS Congress, Amsterdam 2000: ADS40g ,

Highres satellite imaging (within 10 years: footprint improved by factor 10)

PhoWo 2003:Ubiquitous computingAnalogue vs digital image data collection (cameras : DMC UltraCam)Analogue vs. digital image data collection (cameras : DMC, UltraCam)IKONOS, QuickbirdLiDAR DEM Di t ib t d h t t i d t l i A t t d AT i t ti fDistributed photogrammetric data analysis: Automated AT, integration of sensors (GPS/INS), Pixel Factory3D visualization and animation (city models, buildings outside and inside)

30

Space activitiesSpace activities

ESA-ESTEC: Planck, Herschel Gaia 1:1 Mill

NASA‘s Rover SPIRIT

Herschel, Gaia, 1:1 Mill.

Moon Explorer SELENE; JAXA

ESA’s Mars Express

31China‘s Chang‘e-1

D l t f t llitDevelopment of satellite sensors1972 Landsat MSS 80 m GSD

1986 SPOT stereo 10 m

2001 Quickbird stereo 0.6 m

Within 35 years: > Factor 150 in resolution

10.2007: WorldView-1 0.5 m

1.7 days revisit

10.2008: GeoEye -1 0.5 my

32

S i SHRSI Processing: Summary

Georeferencing

• For IKONOS, Quickbird and SPOT-5 bias-corrected RPCs & rigorous models

with same results: planimetry: 0.3 pi height: 0.5 pi

OS/ S C• For ALOS/PRISM and Cartosat-1: planimetry: 0.5 - 0.8 pi height: 0.3 - 0.8 pi

Sufficient for conventional mapping 1: 10 000

DSM generation

• Accuracy (along track): 1-5 pi depending on terrain slope and land covery ( g ) p p g p

• ALOS/PRISM: 2 – 3 pi

• Cartosat: 1 - 3 pi

• But still many large bunders Not ready for mapping 1: 50 000 !

• DSM > DTM reduction not solved yet

33

Radiometric Quality

Z i h i l PRISM Z i hZurich aerial PRISM Zurich

34

A i l S iAerial Sensing

Optical digital cameras (large/medium, oblique)

Laserscanners (LiDAR), GPS/INS

Radar, InSAR

35

M l i l h d

New trend: Oblique imaging

Multiple camera heads

Cam 1Cam 3

Cam 2

Cam 4

© Blom / Pictometry

Cam 2Cam 5

y

36Courtesy: Dr. A. Hellmeier www.rolleimetric.com

Photogrammetry and Remote Sensing:Photogrammetry and Remote Sensing:Turning images into n-dim models (+ semantic info)

E l FiExample: Firenze

37

A REAL-TIME PHOTOGRAMMETRIC MAPPING SYSTEM, Sherman Wu et al.

One Path Photogrammetric Program (OPPP): ProcessesOne-Path Photogrammetric Program (OPPP): Processes automatically DEM, ortho-images, and contour lines on-board and in real-time

38

D l t f lDevelopment of close-range sensors

Low cost cameras

Panoramic cameras

Laserscanners, structured light

3D CCD/CMOS chips

Hybrid systems

Large format cameras:Hasselblad&Phase One:Hasselblad&Phase One: 39 Mpi, 2 sec/image, ca. SFr 40 000

Mobile (Ubiquitous) Photogrammetry

Example: Sony Ericsson K750, 2Mpi camera,

bl t th UBS t

Ladybug26 CCDs

39

bluetooth, UBS, etc. 6 CCDs

30 FPS

4.7 Mpi/frame

5 M i K5 M pi Kamera

video 30 f/sec

GPS

40

8 Gbyte Speicher

Cultural Heritage

Weary Herakles

Alfred Escher

Petroglyphs in Chichictara

41

Bamiyan Buddha

St. Gallen Globus

Fast 3D modeling Templo los Maca ones Campeche Me icoFast 3D modeling – Templo los Macarones, Campeche, Mexico

Recovered camera posesSome of the original frames (total 6 images, 1856x1392)Object dimensions: 3 x 1.5 m

Das Bild kann nicht angezeigt werden. Dieser Computer verfügt möglicherweise über zu wenig Arbeitsspeicher, um das Bild zu öffnen, oder das Bild ist beschädigt. Starten Sie den Computer neu, und öffnen Sie dann erneut die Datei. Wenn weiterhin das rote x angezeigt wird, müssen Sie das Bild möglicherweise löschen und dann erneut einfügen.

Das Bild kann nicht angezeigt werden. Dieser Computer verfügt möglicherweise über zu wenig Arbeitsspeicher, um das Bild zu öffnen, oder das Bild ist beschädigt. Starten Sie den Computer neu, und öffnen Sie dann erneut die Datei. Wenn weiterhin das rote x angezeigt wird, müssen Sie das Bild möglicherweise löschen und dann erneut einfügen.

Shaded 3D model(2M points and 20 000 edges)

Textured 3D model

42

Breuckmann optoTOP-HE system• Accuracy ca. 50 microns• 1.5 days with 67 scans• Each scan 1.3M pointsEach scan 1.3M points

43

44

M ltiM lti hhMultiMulti--sensor approachessensor approachesProject St. Project St. GallenGallen Globe Globe ReplikaReplika

Data acquisition:ca. 460 Scans with stripe projector andca. 250 + 2000 images with digital camerasg g

Data processing:Object-accuracy ~0.8 mmObject accuracy 0.8 mm

45

St G ll Gl b R lik 3D d lSt G ll Gl b R lik 3D d lSt. Gallen Globe Replika, 3D modelSt. Gallen Globe Replika, 3D model

Hybrid (textured) model

Wireframe model

46

Analysis of Interior by X-Rays

• X-rays to analyse the interior of the Globe (construction of axis, materials)

Performed by EMPA (Annex Institute of ETH Zurich for Material Science)• Performed by EMPA (Annex Institute of ETH Zurich for Material Science)

47

Axis and metal parts Material behind the globe surface

Use of hybrid systems: Laserscans and images

3D city modeling: 3D LANDMARKS for car navigation3D city modeling: 3D LANDMARKS for car navigation

Courtesy CyberCity AG

© Harman/Becker, CyberCity

48

3D city modeling: 3D LANDMARKS for car navigation

Courtesy CyberCity AGCourtesy CyberCity AG

49

© Harman/Becker, CyberCity

Forum Pompeji –

3D M d li f l it i3D Modeling of complex site using

multi-sensor, multi-resolution

approaches

FACOLTÀ DI

DISEGNO INDUSTRIALE

REVERSE MODELING+ Dept. INDACO, Politecnico of Milano, Italy

I tit t f G d d Ph t t ETH Z i h S it l d+ Institute of Geodesy and Photogrammetry - ETH Zurich, Switzerland+ Centre for Scientific and Technol. Research – B. Kessler Foundation, Trento, Italy+ Dept. of Applied Sciences, Parthenope University, Naples, Italy

Collaboratori:Laura L. MicoliMichele Russo

Used technologies and multi-resolution data

Sensors Use Quantity Geometric resolution

Texture resolution

Used technologies and multi resolution data

Aerial images

Zeiss RMK A 30/23DSM of the site at low resolution

3

(scale 1:3500)25 cm 5 cm

imagesPictometry Texturing 4 - 15 cm

Leica HDS3000 Modeling of 21 scans 5-20 mm -

Range sensors

Leica HDS3000 gentire Forum

at middle resolution

(400 Mil pts)5 20 mm

Leica HDS600045 scans

(800 Mil pts)5-10 mm -

Canon 10D

(24 mm lens, 6 Mpixel)

Canon 20DModeling of small finds

Terrestrial images

Canon 20D

(20 mm lens, 8Mpixel)

Kodak DCS Pro

(50 mm lens, 12 Mpixel)

small finds, mural

architectural structures,

> 3000 0.5-10 mm 0.2-5 mm

51

(50 mm lens, 12 Mpixel)

Nikon D300

(20 mm lens, 12 MPixel)

ornaments

5

High resolution: Photogrammetry with interactive modelingHigh resolution: Photogrammetry with interactive modeling

• Over 3000 images acquired

dl bl k dj• Bundle block-adjustment

• Interactive modeling to define the main geometric entities

• Geometric resolution (footprint): 0.5-10 mm

52

5

Photogrammetric block for a large structure (temple of Jupiter)

Forum Pompeji - 3D modelForum Pompeji 3D model

53

5

The Sacred Mountain in Social Context:The Sacred Mountain in Social Context: Symbolism and History in Maya Architecture—

Temple 22, Copan HondurasTemple 22, Copan Honduras

54

55

“TEMPLE 22” is known as the “Parthenon of the Maya world” because its remarkable sculpture is in museums around the world

A NEW VISION OF TEMPLE 22 EXISTS THE ANSWER:A NEW VISION OF TEMPLE 22 EXISTS,BUT HOW TO EXPLAIN THIS COMPLEXITY OTHER THAN IN 2D DRAWINGS?

reality-based 3D model and reconstruction of Temple 22, East Court and Principal Group at varying levels ofGroup, at varying levels of detail.

METHODS:1. Photogrammetry and

laserscanning of architecture and sculpture on site, and in warehouses and museums via

BEFORE: Hohmann and Vogrin 1982warehouses and museums via photogrammetry and structured light for reality-based reconstruction.

2. CAD integration and VR simulations for the reconstructed portions. p

3. Develop 3D GIS website for online download and viewing of the model (among other

56AFTER: von Schwerin and Allen 2009

of the model (among other things).

UAV h t t (P l /N P )UAV photogrammetry (Palpa/Nasca, Peru)

3D model

57

Heli flight

St d t j t C tl L d bStudent project: Castle Landenberg

58 58Symbol of Obwalden in the Swiss passport

Landenberg: Flight planning

• Circle with a radius of 25 metersCircle with a radius of 25 meters

• 24 images; every 15°

• Images oblique• Images oblique

• Simulation

59 59

St d t j t C tl L d bStudent project: Castle Landenberg

60

A t S t B1 110Aeroscout Scout B1-110 with aerial laser scanner

61

New Motion Capture standards

Vicon: MX 20+ Motion Capture Camera, Markers

2 Mill. pi, 166 frames/sec, 10 bit

MX Ultranet: 245 cameras simultaneously

Movie Troy (Prad Pitt): y ( )

555 shots, up to 5 performers simult.3 hrs MOCAP data

62

Motion Analysis

The Return of the King: 18 000 virtual effects shotsg

3 500 Linux-based processors

420 people (Weta Digital, Wellington)

Gollum: MoCap, roto-motion, keyframe animation

16 cameras à 1.3 Mpi, 60/120 fr/sec

10 l M C 15 l M Edit10 people MoCap, 15 people MoEdit

63

3D movies (+ TV channel BSkyB, 2010)

Steven Spielberg (TA 4.7.09) on the revolution of the 3D film: p g ( )

“ I find it fascinating how 3D images resemble real life; we also see reality three-dimensionally”see reality three dimensionally

Producer ofProducer of „Transformers: The Revenge of the Fallen“

d

64

and„The Adventures of Tintin“

GeogamesGeogames

• DTM and satellite/aerial images

Microsoft Flight Simulator Xg

• DTM and train-track data

65

Microsoft Train Simulator 2

GeogamesGeogames

• Laserscan to obtain race track geometry

• Motion capture to animate (laser scanned) modeliracing.com – motorsport simluations

66

NHL 2k9 – images by cnet.com

Photogrammetry – a secret technology?

A i ti t I h d i t it D i ti„As a scientist I had a unique opportunity. During my time astronomy reached the market places.“

Galileo Galilei to Andrea Sarti in „The Life of Galilei“, by Bertold Brecht

67

The Market PlaceThe Market Place

2.2009: Teleatlas maps Thailand3.2009: GeoEye-1 for Google Earth8.2009: WorldView-2 arrives at Vandenberg

Air Force Base, Launch 6.10.09

28/05/2009: RapidEye Constellation's 100 DaysRapidEye have collected more than 69 Million square kilometres of at least 80% cloud-free images during their first 100 days of operation.The imaging campaigns were concentrated in Europe, the U.S., Brazil and China.

68

g g p g p , ,More than 36 Million square kilometres of the Earth's surface, or about one quarter of the total landmass were imaged, often several times.

28/05/2009: 3D Models of European CitiesBlom has completed the production of the first 40 high quality 3Dmodels, Blom3D, of European cities. The Blom3D models have been, , pdelivered to Tele Atlas for integration into navigation, LBS and mappingsolutions.Tele Atlas has already announced that they have launched theTele Atlas has already announced that they have launched the photorealistic buildings and blocks created by Blom, as part of their navigation solutions. Their solution, named Advanced City Models, will enhance the user experience of any navigator integrating the solutionenhance the user-experience of any navigator integrating the solution.

69

/ / S f01/06/2009: Large-Scale Mapping for UKThe GeoInformation Group has launched a new mapping programme.UKMap is the UK's first commercially funded, large-scale topographic mapping and address database created completely independently of the Ordnance Survey.

Employing more than 100 people across three continents, UKMap is a five-year programme to map over 500 towns and cities coveringall urban areas with a population greater than 10 000 24 000 kil t th h t10,000 - some 24,000 square kilometres throughout the UK.London is the first complete UKMap city with over 1 700 sqkm and will be available on 1 Sept 20091,700 sqkm and will be available on 1 Sept. 2009.

Geo-enabling Mobile Phones (navigation, maps, 3D models, …)

O.-P. Kallasvuois, President&CEO Nokia, 4.2009: ”We have focused our investments

70

on five primary categories: maps, music, messaging, media and games.”

Changes in society & economyg y yNew environments for R&D, Education and Professional Practice

+ New interest in geosciences

Natural and man-made hazards

+ Communication

Technology & knowledge transfer, data accessibilityTechnology & knowledge transfer, data accessibility

+ Globalization of science

+ Diversification interdisciplinarity+ Diversification, interdisciplinarity

Technology and applications

71

f iImpact on profession(a) Inflation of data

„An image is worth a thousand words“ – What will 100 Mill. images tell us?

Ubiquitous imaging, Cloud-based ComputingP i biliti t ili b hi d d t i iti tProcessing capabilities trailing behind data acquisition rate

ETHZ: Center for Imaging Sciences & Technologies

Google Microsoft street images: Transparent human personality issuesGoogle, Microsoft street images: Transparent human, personality issues

(b) Increased system complexityMore blackboxes or better/more education?More blackboxes or better/more education?

(c) Competition from neighbouring disciplines(c) Competition from neighbouring disciplinesDevelopment of own capabilities and attitudes

(openness flexibility self-assurance)

72

(openness, flexibility, self-assurance)

Maintain depth in research

73

Das Bild kann nicht angezeigt werden. Dieser Computer verfügt möglicherweise über zu wenig Arbeitsspeicher, um das Bild zu öffnen, oder das Bild ist beschädigt. Starten Sie den Computer neu, und öffnen Sie dann erneut die Datei. Wenn weiterhin das rote x angezeigt wird, müssen Sie das Bild möglicherweise löschen und dann erneut einfügen. Das Bild kann nicht angezeigt werden. Dieser Computer verfügt möglicherweise über zu wenig Arbeitsspeicher, um das Bild zu öffnen, oder das Bild ist beschädigt. Starten Sie den Computer neu, und öffnen Sie dann erneut die Datei. Wenn weiterhin das rote x angezeigt wird, müssen Sie das Bild möglicherweise löschen und dann erneut einfügen.

74

Photogrammetry today:

+ From point positioning and 2.5D mapping to an integrated,

unified n-D technology, encompassing satellite, aerial and terrestrial

l fsensor platforms

+ From single sensor/multiple processing instruments to multiple sensors/single processing platform techniquesensors/single processing platform technique

(1) Technologies are converging (satellite, aerial, terrestrial)Processing (almost) platform-independentProcessing (almost) platform-independent

(2) Image understanding is a hard problemFull automation only possible for highly structured images or unstructured productsor unstructured productsRemedy: Multi (hybrid)-data approach (?)

75

Image-, scene understandingRodney Brooks, MIT (The Road to 2030 - A Wild Ride), 2005:

” (For navigation) The robots of tomorrow will need to have the object recognition capabilities of a 2-year old child,…”

Measurement or

76

Measurement or modeling errors ??

Do not forget:

Photogrammetric principles/prioritiesotog a et c p c p es/p o t es

* Sophisticated sensor models, network competence

* Refined measurement algorithms ( precision)

* Redundant data ( reliability)

* Self-diagnosis, quality control

* S t d i f l li bilit* System design for general applicability

* Engineering approach: Testing, validation, robustification

Service to profession !

77

Service to profession !

The Future Al Gore, Vicepresident USA, January 1998:p y

„Working together, we can help solve many of the most pressing problems facing our society, inspiring our children to learn more p g y, p gabout the world around them, and accelerate the growth of a multi-billion dollar industry.“

78